Start here if you're new to NLP. We cover tokenization, parsing, and basic sentiment analysis. You'll build a simple classification system by week four. Most students come in knowing Python basics and leave understanding how language models actually process text.

• Build a working sentiment analyzer
• Process real Twitter data
• Debug common NLP mistakes
• Set up your own development environment

This one gets into transformers and attention mechanisms. You need solid Python and some ML background. By the end you'll fine-tune a pre-trained model for a specific task. It's challenging but most students find the practical focus makes complex concepts click.

• Fine-tune BERT models
• Work with HuggingFace libraries
• Understand attention mechanisms
• Optimize model performance
• Handle real computation constraints

Taking models from notebook to production. We cover deployment, monitoring, and dealing with drift. You'll learn about API design, latency issues, and cost management. This matters if you want to build systems people actually use rather than just experiments.

• Deploy models to AWS/Azure
• Build monitoring dashboards
• Handle model versioning
• Manage inference costs
• Deal with real-world edge cases